level-up
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a process-oriented reflection tool designed to convert weekly friction into concrete improvements through a structured questionnaire. It does not contain any prompt injection, obfuscation, or persistence mechanisms.
- [SAFE]: While the skill reads local files such as
decisions/log.mdand~/.claude/usage.jsonlfor context, it does not perform any network operations, and there is no risk of data exfiltration. - [NO_CODE]: The skill instructions include a mandatory "Hard rule" stating that no code is written by this skill; it only produces a plan for the user to review and implement in a separate session.
- [SAFE]: The skill operates with a 'least privilege' approach by requiring user approval before logging recommendations and strictly limiting its output to planning artifacts.
Audit Metadata